Mapping a locus for alcohol physical dependence and associated withdrawal to a 1.1 Mb interval of mouse chromosome 1 syntenic with human chromosome 1q23.2-23.3

Authors

  • L. Kozell,

    1. Department of Veterans Affairs Medical Center and Portland Alcohol Research Center, Oregon Health & Science University, Portland, OR, and §Roudebush VA Medical Center and Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN, USA
    Search for more papers by this author
  • , J. K. Belknap,

    1. Department of Veterans Affairs Medical Center and Portland Alcohol Research Center, Oregon Health & Science University, Portland, OR, and §Roudebush VA Medical Center and Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN, USA
    Search for more papers by this author
  • , J. R. Hofstetter,

    1. Department of Veterans Affairs Medical Center and Portland Alcohol Research Center, Oregon Health & Science University, Portland, OR, and §Roudebush VA Medical Center and Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN, USA
    Search for more papers by this author
  • , A. Mayeda,

    1. Department of Veterans Affairs Medical Center and Portland Alcohol Research Center, Oregon Health & Science University, Portland, OR, and §Roudebush VA Medical Center and Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN, USA
    Search for more papers by this author
  • and , K. J. Buck ,,

    Corresponding author
    1. Department of Veterans Affairs Medical Center and Portland Alcohol Research Center, Oregon Health & Science University, Portland, OR, and §Roudebush VA Medical Center and Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN, USA
    Search for more papers by this author

*K. J. Buck, PhD, Portland VA Medical Center (mailcode R&D40), 3710 Veterans Hospital Road, Portland, Oregon, OR 97239-3098, USA. E-mail: buckk@ohsu.edu

Abstract

Physiological dependence and associated withdrawal episodes are thought to constitute a motivational force perpetuating continued alcohol use/abuse. Although no animal model duplicates alcoholism, models for specific factors, like the withdrawal syndrome, are useful to identify potential determinants of liability in humans. We previously detected quantitative trait loci (QTLs) with large effects on predisposition to physical dependence and associated withdrawal following chronic or acute alcohol exposure to a large region of chromosome 1 in mice (Alcdp1 and Alcw1, respectively). Here, we provide the first confirmation of Alcw1 in a congenic strain, and, using interval-specific congenic strains, narrow its position to a minimal 1.1 Mb (maximal 1.7 Mb) interval syntenic with human chromosome 1q23.2-23.3. We also report the development of a small donor segment congenic that confirms capture of a gene(s) affecting physical dependence after chronic alcohol exposure within this small interval. This congenic will be invaluable for determining whether this interval harbors a gene(s) involved in additional alcohol responses for which QTLs have been detected on distal chromosome 1, including alcohol consumption, alcohol-conditioned aversion and -induced ataxia. The possibility that this QTL plays an important role in such diverse responses to alcohol makes it an important target. Moreover, human studies have identified markers on chromosome 1q associated with alcoholism, although this association is still suggestive and mapped to a large region. Thus, the fine mapping of this QTL and analyses of the genes within the QTL interval can inform developing models for genetic determinants of alcohol dependence in humans.

Alcoholism and alcohol abuse are complicating factors in most chronic illnesses, and affect up to 30% of Americans (Hasin et al. 2007). Alcoholism is one of the most highly heritable addictive disorders (Goldman et al. 2005). Unfortunately, because of its complex nature, the genetic determinants of alcoholism remain largely unknown, hindering effective treatment and prevention.

Physiological dependence and associated withdrawal episodes are thought to constitute a motivational force that perpetuates continued alcohol use/abuse and contributes to relapse. When alcohol is eliminated (use is discontinued), as its depressant effects disappear, characteristic symptoms of hyperexcitability first wax and then wane, including tremor, autonomic nervous system overactivity and, in extreme cases, convulsions. These withdrawal symptoms define a preexisting state of physiological dependence on alcohol. In alcoholics, seizures, one of the most feared and life-threatening manifestations of the withdrawal syndrome, are known to have a genetic predisposition (Lutz et al. 2006; van Munster et al. 2007) and aggregate in families (Schaumann et al. 1994).

Although no animal model can duplicate alcoholism in humans, models for specific traits, including withdrawal, are useful for identifying potential determinants of susceptibility in humans. Alcohol withdrawal convulsions provide a quantitative index of withdrawal severity in mice (Crabbe et al. 1991; Goldstein & Pal 1971) with a clear genetic contribution to individual differences (Goldstein 1973). Other signs of withdrawal are highly correlated with convulsive activity (Belknap et al. 1987; Kosobud & Crabbe 1986), so common practice is to use convulsive activity as the primary measure of ethanol withdrawal syndrome severity. The most commonly used models of ethanol withdrawal are chronic models, where withdrawal is monitored after mice are continuously exposed to ethanol vapor for 3 days (Terdal & Crabbe 1994) or exposed to a liquid diet containing ethanol for several days (Levental & Tabakoff 1980), and the acute model where withdrawal is monitored hourly for 12 h after a single hypnotic dose of ethanol (Metten et al. 1998). The acute model has the advantage that blood ethanol concentrations (BECs) vary little across genotype, so that monitoring of BECs is not usually necessary. Also, there is no need for pyrazole (alcohol dehydrogenase inhibitor) injections to stabilize BECs as is required in the chronic vapor method (Terdal & Crabbe 1994). The chronic model has the advantage that withdrawal is more intense, allowing for more accurate quantification of individual and genetic differences.

Genome-wide evaluations of withdrawal severity, indexed using the handling-induced convulsion (HIC), identified quantitative trait loci (QTLs) on chromosome 1 with large effects on withdrawal following chronic and acute alcohol exposure (Alcdp1 and Alcw1, respectively; Buck et al. 2002, 1997; http://www.informatics.jax.org/searches/marker_report.cgi), as well as pentobarbital withdrawal (Pbw1; Buck et al. 1999). These QTLs identified the approximate genomic location of a gene(s) affecting withdrawal but were initially mapped to large confidence intervals containing hundreds of genes, any one (or more) of which could be responsible for the QTL associations. High-resolution mapping is therefore crucial to progress toward identification of the gene(s) that underlie QTL phenotypic effects and potential pleiotropic effects on additional behaviors. One of the most powerful strategies to precisely fine map a QTL employs interval-specific congenic strains (ISCS) (Cicila et al. 2001; Darvasi 1997, 1998; Fehr et al. 2002; Lyons et al. 2000; Shirley et al. 2004). Here, we used this strategy to fine map QTLs affecting physiological dependence following chronic and acute alcohol exposure to a small interval of murine chromosome 1 syntenic with human chromosome 1q23.2-23.3. Markers on human chromosome 1q are associated with alcoholism (Aragaki et al. 1999; Dick et al. 2002; Guerrini et al. 2005; Turecki et al. 1999), although these are still suggestive associations and map to large regions. We, therefore, expect that the present fine mapping results and analyses of the genes within this QTL interval (Denmark & Buck 2008) will inform developing models for gene influences on alcohol physical dependence and alcoholism in humans.

Materials and methods

Animals

Two mouse strains, DBA/2J (D2) and C57BL/6J (B6), were used to develop congenic strain mice. The D2.B6-D1Mit206 (D2.B6) congenic strain was developed in our colony at the Veterinary Medical Unit of the Portland Veterans Affairs (VA) Medical Center. B6, D2 and B6.D2-Mtv7a/Tycongenic (Taylor & Frankel 1993) strain mice were purchased from the Jackson Laboratory and bred in our colony. Four interval-specific congenic strains derived from the B6.D2-Mtv (B6.D2) congenic strain were developed in our colony or by Drs John Hofstetter and Aimee Mayeda at the Indianapolis VA Medical Center. All of the animals used for behavioral testing were bred in our colony. A total of 926 mice were behaviorally tested over a period of 2 years, including comparisons of 758 congenic, 142 B6, and 26 D2 strain mice. Mice were group-housed 2–5 per cage by strain and sex. Mouse chow (Purina LabDiet 5001; Purina Mills International, St Louis MO, USA) and water were available ad libitum. Procedure and colony rooms were kept at a temperature of 21 ± 1°C. Lights were on in the colony from 0600 to 1800 h, and behavioral testing was initiated between 0700 and 0800 h. All procedures were approved by the VA and Oregon Health & Science University Institutional Animal Care and Use Committees in accordance with United States Department of Agriculture and United States Public Health Service guidelines.

Development of a series of interval-specific congenic strains

Previously, we showed that a QTL affecting chronic ethanol withdrawal severity was captured within the introgressed interval of the chromosome 1 congenic strain B6.D2-Mtv (Buck et al. 2002). We tested B6.D2 and an approximately reciprocal congenic, D2.B6, for acute ethanol withdrawal severity. Genotypic analysis delimited their minimal (and maximal) introgressed intervals of chromosome 1 as follows: D2.B6, 152.2–176.5 Mb (maximal 151.6–177.5 Mb); B6.D2 172.9–188.0 Mb (maximal 172.3–189.3 Mb) (unpublished results). Because B6.D2 congenics had a smaller introgressed region, this congenic was used as our point of departure to generate a series of ISCSs for higher resolution QTL mapping. Mice from the B6.D2 congenic mice were crossed to B6 mice to yield F1 (B6.D2 X B6) animals, which were backcrossed to B6 mice. Individual progeny were genotyped using D1Mit and single nucleotide polymorphism (SNP) markers within or flanking the acute and chronic alcohol withdrawal QTLs on chromosome 1 (referred to as Alcw1 and Alcdp1; Buck et al. 1997, 2002, to identify recombinant mice and define the boundaries of the introgressed intervals. Individual recombinant mice were backcrossed to B6 mice, resulting in multiple offspring with the same recombination (referred to as an interval-specific congenic line). At the same time that recombinations in the prior generation were being replicated, additional recombinants were sought in subsequent backcross generations to develop smaller QTL intervals and replicated as needed. For four lines, final intercrosses were performed to isolate the donor homozygotes, which constituted the four finished ISCS (i.e. R4, R6, R7 and R8) that, in addition to the starting congenic, were used for phenotypic analysis of alcohol withdrawal severity. We tested for QTL capture by phenotypic comparisons of congenic and background strain (B6) mice.

Alcohol withdrawal phenotypic analyses

Physiological dependence is operationally defined as the manifestation of physical disturbances (withdrawal syndrome) after alcohol administration is suspended. Genetic variation in alcohol withdrawal severity was examined by monitoring HICs associated with withdrawal, which is a sensitive index of alcohol withdrawal severity (Crabbe et al. 1991; Goldstein & Pal 1971). Details of the HIC scoring system are given in Table 1. McQuarrie and Fingl (1958) first showed a state of withdrawal central nervous system (CNS) hyperexcitability after acute alcohol administration. Details of the acute ethanol withdrawal procedure have been published (Metten et al. 1998). Individual mice and different inbred strains can differ in baseline (pre-ethanol) HIC scores. Therefore, to assess acute ethanol withdrawal severity, mice were scored twice for baseline (pre-ethanol) HICs 20 min apart, followed by a single sedative–hypnotic dose of ethanol (4 g/kg, i.p. in 20% w/v in saline) and were scored hourly between 2 and 12 h post-ethanol. To create an index of ethanol withdrawal that is independent of individual differences in baseline HIC scores and reflect differences in withdrawal convulsion severity, post-ethanol HIC scores were corrected for the individual’s average predrug (baseline) HIC score as in previous work (Metten et al. 1998). The acute ethanol withdrawal was indexed as the area under the curve (AUC; i.e. the sum of the post-ethanol HIC scores) over the full time–course post-ethanol. Individual ethanol withdrawal severity scores correspond to these AUC values.

Table 1.  HIC rating scale
SymptomScore
No convulsion or facial grimace after gentle 180° spin0
A facial grimace is seen after gentle 180° spin1
No convulsion when lifted by the tail, but a tonic convulsion is elicited by a gentle 180° spin2
Tonic-clonic convulsion after a gentle 180° spin3
Tonic convulsion upon lifting by the tail4
Tonic-clonic convulsion when lifted by the tail, often with the onset delayed up to 1–2 seconds5
Severe, tonic-clonic convulsion when lifted by the tail, with a quick onset and long duration, often continuing for several seconds after the mouse is released6
Severe, tonic-clonic convulsion elicited prior to lifting by the tail, with a quick onset and long duration, often continuing for several seconds after the mouse is released7

Details of the chronic inhalation exposure method used to induce ethanol physical dependence have been published and involve a standard paradigm in which adult mice are continuously exposed to ethanol vapor for 72 h (Terdal & Crabbe 1994). Briefly, mice were weighed and scored twice for baseline HICs, about 20 min apart prior to receiving either saline (air–pyrazole control group) or a loading dose of ethanol (ethanol-exposed group; 1.5 g/kg ethanol, i.p., 20% v/v in saline). In addition, all the mice received daily injections of pyrazole–HCl (68.1 mg/kg, i.p.; an alcohol dehydrogenase inhibitor) to stabilize blood and brain ethanol levels. Levels of ethanol in vapor (typically 6–8 mg ethanol/l of air) were selected to achieve approximately equal blood ethanol levels across strains. After 24 and 48 h of ethanol exposure, blood samples were obtained from 20 mice, which served as an additional check on the efficacy of the inhalation procedures in each pass and allowed minor adjustments to the ethanol flow rates to maintain BECs near the desired blood level (approximately 1.5 mg/ml). About 20 μl blood samples were drawn from the end of the nicked tail with a capillary tube. At 72 h, all mice were removed from the inhalation chambers and immediately tested for HIC severity. Next, blood samples were drawn from the ethanol-exposed mice, and control animals had their tails nicked but no blood was collected. Chronic ethanol withdrawal intensity was monitored using HICs starting immediately after removal from the inhalation chambers, at hourly intervals for 10 h, and at 24 and 25 h following removal from the chamber. Chronic ethanol withdrawal severity was calculated in a similar manner as acute withdrawal severity. First, individual hourly HICs were corrected for baseline HIC scores, then chronic withdrawal was indexed as the AUC (i.e. the sum of the postethanol scores) over the 25-h time–course. To correct for any effect of the daily pyrazole treatment on HIC scores, the AUC25 for each animal was also corrected for the average air–pyrazole withdrawal for each genotype.

Pentobarbital withdrawal phenotypic analyses

R8 congenic and background strain mice were compared for their pentobarbital withdrawal severities. As in previous work (Buck et al. 1999), adult mice were scored twice for baseline HICs immediately before administration of pentobarbital (60 mg/kg, i.p.) and then hourly between 1 and 8 h post-pentobarbital administration. To create an index of pentobarbital withdrawal that is independent of individual differences in baseline HIC scores and reflect differences in withdrawal convulsion severity, postpentobarbital HIC scores were corrected for the individual’s average baseline HIC score as in previous work, and pentobarbital withdrawal was indexed as the AUC (i.e. the sum of postpentobarbital scores) over the entire time–course. Individual pentobarbital withdrawal severity scores correspond to these AUC values.

Genotypic analysis

DNA for genotyping was extracted from tail biopsy or ear punch tissue using the Puregene® DNA isolation kit (Gentra Biosystems, Minneapolis, MN, USA) according to the manufacturer’s instructions. polymerase chain reaction amplification and gel electrophoresis was performed as in previous work (Fehr et al. 2002) using markers from the D1Mit series of SSLP and SNP markers for mouse chromosome 1 (www.informatics.jax.org).

Data analysis

For B6.D2-Mtv and D2.B6-D1Mit206 congenic vs. appropriate background strain comparisons for acute ethanol withdrawal, the data were found to be normally distributed based on a nonsignificant Shapiro–Wilks tests, and were analyzed using an analysis of variance (anova) followed by post hoc (Tukey) analyses. For all other behavioral comparisons, the data were not normally distributed by the same test and were analyzed using nonparametric Kruskal–Wallis anova on ranks, which generates a Kruskal–Wallis test statistic and corresponding P value; when significance was indicated (α < 0.05), this analysis was followed by a Kolmogorov–Smirnov analysis (Systat 11; Systat Software Inc., San Jose, CA, USA).

Results

Capture of an acute alcohol withdrawal QTL (Alcw1) in chromosome 1 congenic strains

We began by testing two approximately reciprocal congenic strains (D2.B6 and B6.D2) and appropriate background strain mice (D2 and B6) for acute ethanol withdrawal indexed using the HIC. There was a significant difference in acute alcohol withdrawal severity because of strain in the B6.D2 congenic vs. B6 background strain comparison (P = 1.6 × 10−7), and also in the D2.B6 congenic vs. D2 background strain comparison (P = 2.4 × 10−11). There was also a significant main effect of sex in the D2.B6 vs. D2 strain comparison (P = 0.0012), with male mice exhibiting more severe withdrawal than female mice in both strains. However, no strain × sex interaction was apparent for either congenic vs. background strain comparison (P = 0.38 and 0.27, respectively). Figure 1a illustrates the HIC time–course associated with acute alcohol withdrawal for D2.B6 and B6.D2 congenics vs. the appropriate background strains (D2 and B6, respectively). Figure 1b summarizes these data as the AUC and illustrates that D2.B6 congenic mice exhibited significantly less severe alcohol withdrawal than D2 background strain mice (P = 2 × 10−9), and that the approximately reciprocal B6.D2 congenic strain had more severe ethanol withdrawal than B6 background strain mice (P = 1.4 × 10−7). Taken together with our previous work (Buck et al. 1997), these congenic data confirm that Alcw1 is a highly significant QTL (combined P = 1 × 10−19, LOD = 11.2), which exceeds the guidelines recommended by Lander and Kruglyak (1995) for highly significant linkage.

Figure 1.

Time–course for acute ethanol withdrawal in D2.B6 and B6.D2 congenic and background strain mice. (a) Acute ethanol withdrawal was indexed using the HIC in B6.D2 congenic (closed circles) and B6 background strain mice (closed squares); and in D2.B6 congenic (D2.B6-D1Mit206; open circles) and D2 background strain mice (open squares). The mice were scored twice for baseline HICs immediately before administration of 4 g/kg ethanol (arrow indicates ethanol injection at time 0), and hourly between 1 and 12 h post-ethanol administration. After 4–5 h, convulsion scores increase above baseline indicating a state of withdrawal hyperexcitability, which peaks about 6–8 h post-ethanol administration. Data represent the strain mean ± SEM (n = 63, 38, 34 and 26 mice per strain, respectively). (b) Ethanol withdrawal severity (corrected AUC for 12 h post-ethanol, mean ± SEM) for B6.D2 congenic and B6 background strain mice are shown as well as D2.B6 congenic and D2 background strain mice. Ethanol withdrawal was significantly more severe in the B6.D2 congenic than in B6 background strain mice (F1,99 = 32.3, P = 1.4 × 10−7; n = 63 and 38, respectively). Consistent with these results, ethanol withdrawal was significantly less severe in the D2.B6 congenic than in D2 background strain mice (F1,58 = 44.1, P = 1.2 × 10−8; n = 34 and 26, respectively). As expected, B6, B6.D2 and D2.B6 congenic mice all exhibited much less severe acute alcohol withdrawal than the D2 strain, which is well-documented for its susceptibility to withdrawal after chronic and acute ethanol exposure (Kakihana 1979; Crabbe 1983; Metten & Crabbe 1994). **P < 0.05 (two-tailed).

Our data confirm that Alcw1’s influence on acute alcohol withdrawal is detected on both the D2 and B6 genetic backgrounds. Genotypic analysis of the D2.B6 and B6.D2 congenic strains delimited the minimal (and maximal) introgressed intervals of chromosome 1 as follows: D2.B6, 152.2–176.5 Mb (maximal 151.6–177.5 Mb) and B6.D2, 172.9–188.0 Mb (maximal 172.3–189.3 Mb). The introgressed interval was substantially smaller in B6.D2, so it was used as the starting congenic for the development of ISCSs.

Fine mapping of Alcw1 using a series of interval-specific congenic strains

We began by developing four ISCSs (R4, R6, R7 and R8) derived from the B6.D2 congenic strain (Fig. 2a). The results of phenotypic comparisons of these four ISCSs, B6.D2 congenic and B6 background strain mice are shown in Fig. 2b. Acute alcohol withdrawal severity was significantly different among the strains (H5,719 = 85.2, P < 1 × 10−15). There was no significant main effect of sex (P = 0.44) or strain × sex interaction (P = 0.74), so the data were collapsed across both sexes. R4 mice exhibited significantly more severe withdrawal than B6 background strain mice (P = 6.0 × 10−8), indicating that the D2 allele for the Alcw1 gene(s) within the introgressed interval (spanning D1Mi113 and D1Mit405) influences alcohol withdrawal severity. The QTL effect was also evident in R6 (P = 0.001) and R8 (P = 6.4 × 10−5) congenic animals. R6 contains a 4.4 Mb introgressed region spanning D1Mit113 and D1Mit541. The minimal R8 introgressed interval is 1.1 Mb (172.9–174.0 Mb, spanning SNPs rs32454955 and rs31579250), and the maximal 1.7 Mb (172.3–174.0 Mb) introgressed interval includes a 0.6 Mb proximal boundary region. The proximal boundary is highly identical by descent in B6 and D2-derived mice and, to date, none of the potential genetic markers within this boundary has proven useful for finer mapping. In contrast, R7 (with an introgressed region spanning D1Mit273 and D1Mit221) did not differ from B6 background strain mice in ethanol withdrawal liability (P = 0.98). R4 and B6.D2 did not differ in their withdrawal severities (P = 0.50) nor did R4, R6 and R8 differ in their withdrawal severities (all P > 0.5). These results suggest that Alcw1 is entirely captured by R8, which is consistent with our results using the chronic model. These results substantially reduce the size of the Alcw1 interval from its original 50 cM (approximately 100 Mb) 1 LOD confidence region (Buck et al. 1997), and substantially reduce the number of genes remaining within the Alcw1 interval (Denmark & Buck 2008).

Figure 2.

Fine mapping of an acute alcohol withdrawal QTL on chromosome 1 (Alcw1) using ISCSs. (a) In addition to the starting B6.D2 congenic strain, four recombinant ISCSs (R4, R6, R7 and R8) were developed and tested to attain higher resolution QTL mapping. The genetic markers examined to establish the congenic interval boundaries are indicated and their locations are given (in Mb). For each congenic strain, the donor D2D2 segment is shown in gray; chromosomal regions homozygous for the B6 background strain allele are shown in black and the boundaries between the B6 and D2 regions are shown in white. Congenics that showed significantly more severe withdrawal than background strain mice, thus showing capture of Alcw1 within the introgressed interval, are noted. (b) Acute ethanol withdrawal severity (mean ± SEM) for the B6.D2, four ISCSs, and B6 background strain mice are shown, and was significantly different across strains (H[df=7] = 85.2, P < 1 × 10−15). Ethanol withdrawal was significantly more severe in B6.D2 (P < 1 × 10−9, n = 89), R4 (P = 6 × 10−8, n = 194), R6 (P = 1.4 × 10−3, n = 140) and R8 (P = 6.4 × 10−5, n = 119) mice than in background strain mice (n = 126). In contrast, ethanol withdrawal severity did not differ from background strain mice for R7 (P = 0.98, n = 51). **P < 0.05 (two-tailed). Taken together, these data narrow Alcw1 to a small interval corresponding to the R8 introgressed interval. The minimal introgressed interval is 1.1 Mb (172.9–174.0 Mb). The maximal Alcw1 interval (172.3–174.0 Mb) includes a 0.6 Mb proximal boundary region.

Finally, although R4, R6 and R8 mice did not differ in their withdrawal severities, R6 and R8 mice did show less severe withdrawal than B6.D2 mice (P = 0.008 and 0.014, respectively). Thus, the possibility remains that an additional, more distal QTL also influences acute ethanol withdrawal, and may overlap more distal QTLs identified for acute pentobarbital withdrawal (L. Kozell, N. Walter, K. Wickman & K. Buck, unpublished data) and/or maximal electroshock seizure threshold (Ferraro et al. 2007).

R8 also captures a QTL for physical dependence and withdrawal following chronic ethanol exposure (Alcdp1)

We also tested B6.D2, R4, R7 and R8 and B6 background strain mice for chronic ethanol withdrawal severity. There were no significant BEC differences among the B6.D2, R4, R8 and B6 strains (BEC values ± SEM = 1.2 ± 0.1, 1.3 ± 0.1, 1.3 ± 0.1 and 1.3 ± 0.1 mg ethanol/ml blood, respectively), while R7 mice tended to have somewhat higher BEC values (1.5 ± 0.1 mg/ml) (H4,147 = 5.2, P = 0.27).

Chronic ethanol withdrawal severity can be indexed as AUC for the 25 h examined postethanol (AUC25), or peak withdrawal averaged across the peak HIC score and the HIC values for the two flanking time-points (PEAK). Analyses using both indexes clearly showed that B6.D2, R4 and R8 congenic strains showed more severe chronic ethanol withdrawal than background strain mice. Using AUC25, a significant difference in chronic alcohol withdrawal severity was apparent across strains (H4,147 = 10.6, P = 0.03; Fig. 3), with B6.D2, R4 and R8 congenic strains exhibiting significantly more severe chronic alcohol withdrawal than B6 background strain mice (P = 0.044, 0.041 and 0.048, respectively). In contrast, R7 and background strain mice did not differ in their withdrawal severities (P = 0.62). Using PEAK, there was also a significant difference in chronic alcohol withdrawal severity because of the strain (H4,147 = 14.6, P = 0.006), with B6.D2, R4 and R8 congenic strains exhibiting significantly greater peak withdrawal than background strain mice (P = 0.05, 0.002 and 0.014, respectively). In contrast, R7 and B6 mice did not differ in their withdrawal severities, despite R7 mice showing a trend for higher BEC levels than the B6 mice (1.5 ± 0.1 vs. 1.3 ± 0.1 mg/ml, respectively).

Figure 3.

Small donor segment congenic (R8) mice show significantly more severe withdrawal than physically dependent background strain mice after chronic ethanol exposure. Chronic ethanol withdrawal severity (mean AUC25 ± SEM) for the B6.D2, three ISCSs (R4, R6 and R8), and B6 background strain mice are shown, and was significantly different across strains (H4,147 = 10.6, P = 0.03). Ethanol withdrawal was more severe in B6.D2 (P = 0.044, n = 14), R4 (P = 0.041, n = 73) and R8 (P = 0.048, n = 33) mice than in B6 background strain mice (n = 16). In contrast, chronic ethanol withdrawal severity did not differ from background strain mice for R7 (P = 0.62, n = 11). **P < 0.05 (two-tailed).

Phenotypic testing of R8 indicates that Alcdp1/Alcw1 does not influence barbiturate withdrawal severity

We previously mapped a QTL for barbiturate (pentobarbital) withdrawal to distal chromosome 1 (Buck et al. 1999). We were surprised to observe that R8 and B6 background strains both showed negligible pentobarbital withdrawal severities (AUC = 0.72 ± 0.10 and 0.75 ± 0.13, respectively), rather than R8 showing more severe withdrawal from pentobarbital as well as ethanol. Although Pbw1 is located within the B6.D2 introgressed interval, it maps more distally to a 0.4 Mb interval captured by R6 but excluded by R8 (Kozell, L., Walter, N., Wickman, K. & Buck, K unpublished data). Thus, Alcw1 is the first confirmed fine mapped QTL for alcohol withdrawal that has not been found to influence barbiturate withdrawal.

Discussion

The present studies were carried out in order to confirm and fine map QTLs on chromosome 1 affecting withdrawal following acute and chronic alcohol exposure. This is a crucial step to progress toward identifying the gene(s) that underlies the QTL effects on these phenotypes. Our results reduce the Alcdp1/Alcw1 region to a minimal 1.1 Mb interval of distal chromosome 1. This interval contains 40 coding genes, a subset of which show validated genotype-dependent transcript expression and/or nonsynonymous coding sequence variation that may underlie the influence of Alcdp1/Alcw1 on ethanol physical dependence and associated withdrawal (Denmark & Buck 2008).

We also report the development of a panel of ISCSs, which will be invaluable to assess potential pleiotropic effects of genes for which QTLs were detected on distal chromosome 1. In addition to Alcdp1 and Alcw1, QTLs are detected on distal chromosome 1 for a variety of behavioral responses to alcohol, including alcohol-conditioned aversion (Risinger & Cunningham 1998), alcohol preference drinking (Tarantino et al. 1998), alcohol-induced locomotor activation (Demarest et al. 1999) and hypothermia (Crabbe et al. 1994). This convergence of alcohol-related QTLs makes it tempting to speculate that the gene(s) underlying the influence of Alcdp1/Alcw1 on alcohol withdrawal may have pleiotropic effects on other behavioral responses to alcohol, and shows the cumulative power of QTL mapping to detect multiple potential effects of what could be the same gene. More definitive confirmation of this hypothesis will require testing the small donor segment congenic strain (R8) for other additional behavioral responses to ethanol to determine whether the congenic vs. background differences are congruent with those reported here for ethanol withdrawal severity. The possibility that Alcdp1/Alcw1 plays an important role in such diverse responses to alcohol makes it an important target for further investigations.

A QTL with a large effect on pentobarbital withdrawal severity has also been confirmed on distal chromosome 1 (Buck et al. 1999) in the same broad region as previously reported for the alcohol withdrawal QTLs. However, in the light of the increase in mapping resolution gained from the present study of alcohol withdrawal QTLs, it appears that Pbw1 lies outside the narrowed interval for Alcdp1/Alcw1. In contrast to results for acute and chronic alcohol withdrawal, R8 and B6 background strain mice did not differ in their pentobarbital withdrawal severities, indicating that Alcdp1/Alcw1 does not influence pentobarbital withdrawal convulsion severity. Pbw1 is localized within the starting B6.D2-Mtv introgressed interval, but is located more distally within a 0.4-Mb interval that excludes the R8 introgressed interval (Kozell, L., Walter, N., Wickman, K. & Buck, K unpublished data). Although our results provide no direct evidence for a second alcohol withdrawal QTL within the Pbw1 interval, the possibility remains that an additional more distal QTL also influences ethanol withdrawal, and may overlap more distal QTLs identified for pentobarbital withdrawal (Kozell, L., Walter, N., Wickman, K. & Buck, K unpublished data) and maximal electroshock seizure threshold (Ferraro et al. 2007). More definitive evidence for or against an additional more distal alcohol withdrawal QTL will likely require analyses using interval-specific D2.B6 congenic strains or gene-targeted animal models under development. However, our results clearly exclude Pbw1 from the Alcdp1/Alcw1 interval. This can provide an important clue so as to the identity of the gene(s) that underlies the influence of the Alcdp1/Alcw1 gene(s) on alcohol physical dependence and associated withdrawal (Denmark & Buck 2008). In contrast, fine mapping of alcohol and pentobarbital withdrawal QTLs on chromosome 4 implicated the same 1.8 Mb interval and gene for both sedative–hypnotic drugs (Shirley et al. 2004). Thus, Alcdp1/Alcw1 is the first confirmed fine mapped locus for alcohol withdrawal that does not influence barbiturate withdrawal HICs.

QTLs for other CNS hyperexcitability states are also detected on distal chromosome 1. QTLs for kainic acid and pentylenetetrazol-induced seizures have been detected (Ferraro et al. 1997), but are mapped to relatively large regions. A QTL analysis for maximal electroshock seizure threshold also identified a QTL on chromosome 1 (Ferraro et al. 2004), but analyses using a bacterial artificial chromosome localize this QTL to the Atp1a4-Atp1a2-Igsf8-Kcnj9-Kcnj10-Pigm region (Ferraro et al. 2007). These six genes are not within the Alcdp1/Alcw1 interval, but are contained within the more distal Pbw1 interval (Kozell, L., Walter, N., Wickman, K. & Buck, K unpublished data), suggesting that the latter may have effects that extend beyond drug withdrawal. Several other mouse models of CNS hyperexcitability have not detected QTLs on chromosome 1, in populations derived from the B6 and D2 strains, including audiogenic seizures (Frankel et al. 1995; Neumann & Collins 1991), seizures induced by a β-carboline or glycine (Martin et al. 1995), and polygenic models of epilepsy (Frankel et al. 1994, 1995). The mechanisms underlying CNS hyperexcitability remain to be elucidated, and some influential genes may indeed have more general effects on hyperexcitability, while other influential genes may be more specific in their effects. Evidence to date suggests that Alcdp1/Alcw1 does not influence other hyperexcitability states like pentobarbital withdrawal convulsions and maximal electroshock seizure threshold, whereas a more distal locus may indeed play an important role in regulating diverse CNS excitability states for which QTLs have been detected on chromosome 1 (Kozell, L., Walter, N., Wickman, K. & Buck, K unpublished data; Ferraro et al. 2007).

The fine mapped Alcdp1/Alcw1 interval is syntenic with human chromosome 1q23.2-23.3. Notably, at least four studies (Aragaki et al. 1999; Dick et al. 2002; Guerrini et al. 2005; Turecki et al. 1999) identified markers on the long (q) arm of human chromosome 1 associated with alcoholism, although these are still suggestive associations and are mapped to large chromosomal regions. Additionally, the Collaborative Study on the Genetics of Alcoholism (COGA) identified markers on chromosome 1q associated with suicidal behavior in multiplex alcohol-dependent families (Hesselbrock et al. 2004), suggesting that a gene or genes in this region may be involved in alcoholism and suicidal behavior, which show high comorbidity. Therefore, we expect that the high-resolution mapping of Alcdp1/Alcw1 reported here, together with analyses of the genes within this QTL interval (Denmark & Buck 2008), will inform future human studies of gene influences on alcohol physical dependence and alcoholism.

Acknowledgments

The authors gratefully acknowledge Drs John Crabbe and Pamela Metten for helpful discussions on this project, and Nikki Walter, Shyla Myrick and Sarah Alexander for excellent technical assistance. This work was supported by three Merit Review Awards from the Department of Veterans Affairs and by National Institutes of Health grants AA011114, AA10760, AA06243 and DA05228.

Ancillary